Prasad Wimalaratne

A Social Media mHealth Solution to Address the Needs of Dengue Prevention and Management in Sri Lanka

Background Sri Lanka has witnessed a series of dengue epidemics over the past five years, with the western province, home to the political capital of Colombo, bearing more than half of the dengue burden. Existing dengue monitoring prevention programs are exhausted as public health inspectors (PHIs) cope with increasing workloads and paper-based modes of surveillance and education, characterizing a reactive system unable to cope with the enormity of the problem. On the other hand, the unprecedented proliferation and affordability of mobile phones since 2009 and a supportive political climate have thus far remained unexploited for the use of mobile-based interventions for dengue management. Objective To conduct a needs assessment of PHIs in Colombo with respect to their dengue-related tasks and develop a new mobile-based system to address these needs while strengthening existing systems. Methods One-on-one in-depth interviews were conducted with 29 PHIs to a) gain a nuanced, in-depth understanding of the current state of surveillance practices, b) understand the logistical, technological and social challenges they confront, and c) identify opportunities for mobile-based interventions. Quantitative analysis included simple descriptive statistics while qualitative analysis comprised textual analysis of 209 pages of transcripts (or nearly 600 minutes of conversations) using grounded theory approaches. Results Current paper-based data collection practices for dengue surveillance involved a circuitous, time consuming process that could take between 7-10 days to officially report and record a single case. PHIs confronted challenges in terms of unreliable, standalone GIS devices, delays in registering mosquito breeding sites and lack of engagement from communities while delivering dengue education. These findings, in concert with a high motivation to use mobile-based systems, informed the development of Mo-Buzz, a mobile-based system that integrates three components – digitized surveillance, dynamic disease mapping and digitized dengue education – on a common platform. The system was developed through an iterative, evolutionary, collaborative process, consistent with the Spiral model of software development and is currently being used by all 55 PHIs in the CMC system. Conclusions Given the entrenched nature of existing paper-based systems in PHIs’ work habits, we expect a gradual adoption curve for Mo-Buzz in the future. Equally, we expect variable adoption of the system with respect to its specific components, and specific PHI sub-groups (younger versus older). The Mo-Buzz intervention is a response to multiple calls by the global mHealth community for collaborations in the area of mobile interventions for global health. Our experience revealed that the benefits of this paradigm lies in alleviating country-specific public health challenges through a commonly shared understanding of cultural mores, and sharing of knowledge and technologies. We call upon future researchers to further dissect the applicability of the Spiral Model of software development to mHealth interventions and contribute to the mHealth evidence debate from theoretical and applied perspectives.

Social Media-Based Civic Engagement Solutions for Dengue Prevention in Sri Lanka: Results of Receptivity Assessment.

This article focuses on a novel social media-based system that addresses dengue prevention through an integration of three components: predictive surveillance, civic engagement and health education. The aim was to conduct a potential receptivity assessment of this system among smartphone users in the city of Colombo, the epicenter of the dengue epidemic in the island country of Sri Lanka. Grounded in Protection Motivation Theory (PMT) and using a convenience sampling approach, the cross-sectional survey assessed perceived severity (PSe), perceived susceptibility (PSu), perceived response efficacy (PRE), perceived self-efficacy (PSE) and intention-to-use (IU) among 513 individuals. The overall receptivity to the system was high with a score of >4.00 on a five-point scale. Participants belonging to younger, better educated and higher income groups reported significantly better perceptions of the efficaciousness of the system, were confident in their ability to use the system, and planned to use it in the future. PMT variables contributed significantly to regression models predicting IU. We concluded that a social media-based system for dengue prevention will be positively received among Colombo residents and a targeted, strategic health communication effort to raise dengue-related threat perceptions will be needed to encourage greater adoption and use of the system.

Simulated hornet attack on crowd in a virtual environment

Simulation of natural phenomena in a virtual environment is an intriguing and mathematically complex open research area. The main contribution of this research is to simulate the behaviour of hornets by exploring the dynamic behaviour of swarm of simple agents. Swarm has been formed with collection of simple agents where each agents acts relatively independent from other agents in the environment. Here the swarm intelligence is gained through Particle Swarm Optimization(PSO) algorithm combining with principles of flocking behaviour which can be used in entertainment industry as well as in military applications.

Towards Communication with Virtual Humans

Natural interaction between application user of a Virtual Environment (VE) and autonomous characters is a key challenge in enhancing the realism in virtual environments. Traditional interaction methods with autonomous characters such as virtual humans using keyboard and mouse do not provide an intuitive user experience. This paper presents an approach that enables user to communicate and control virtual humans in an intuitive manner compared to traditional approaches. Such a metaphors would significantly benefit the application domains such as training simulations, 3D social media networks and computer gaming. The approach uses head direction of user and voice commands in grabbing attention of a Virtual Human and initiating a conversation. The prototype implemented uses computer vision based head tracking approach with one web camera and four LEDs. The user study results show that the approach provides an enhanced user experience in the VE.

On the detection and analysis of software security vulnerabilities

Software security is considered as an important issue in the modern software development practices. According to the literature, the benefits of the existing static analysis tools in isolation for security specific analysis of multiple large-scale software systems is highly questionable. In particular, the existing static analysis tools are not facilitating different analysis directions such as classification of vulnerabilities and analysis of multiple projects for multiple versions. To tackle this problem, this paper presents a conceptual framework and a proof-of-concept implementation to integrate the outputs of multiple static analysis tools into one place to perform a wide range of software security analyses. The rich features of the framework have been demonstrated by using two different application scenarios. We used a case study to evaluate the usefulness and the extensibility of our framework in performing in-depth security analysis.

Sensor fusion for visually impaired navigation in constrained spaces

This work presents a multi-sensor fusion approach for an electronic navigation aid for the blind and visually impaired persons. This approach proposes to intelligently fuse the surrounding information senses via ultrasonic sensors and vision sensors. The intelligent component of the prototype serves in several facets including object detection and recognition. Extended Kalman filter is used to fuse the data emerging from homogeneous sensors and rule-based fusion is used to fuse the data from heterogeneous sensors. Feedback is provided via tactile and audio feedback. Critical obstacles of blind navigation like staircases are recognized by the Hough line detection in image processing. Rotations, which occur due to the body movements in the camera, correct using the fusion of data obtain by the inertial measurement unit which is connected to the camera. The results of the evaluations proved that the use of fusion of multiple homogeneous sensors improve the detection of a particular obstacle and fusion of vision and ultrasonic sensors improve the object detection identification of the obstacles. The current status of the work and the future developments are presented in this paper.

Towards Virtual Therapy for Alcoholic Depression

Virtual Reality Exposure Therapy (VRET) is an alternative to standard graded exposure, in treating alcoholic depression, where participants are exposed in to 3D environments, a bar and a beach, with spatial, narrative and physical immersion since standard therapies present an unsolvable dilemma for some persons. Cue exposure of virtual bar with coping skills decreases drinking and elicits craving in real situations. Beach as a relaxation technique gives positive changes physically and mentally and reduces depression. Skin resistance as a measure of emotional state is used for monitoring relaxation level through the environments. Alcohol craving and attention increased across bar and reduced drastically in beach on severe drinkers. Participants perceived high levels of relaxation and a distraction from alcohol craving in beach. VRET presents a mind distraction to beach when experiences alcohol, controls drinking and reduces depressed mood. Hence, VRET is a promising new medium for treating alcoholic depression.

Towards Realism: Selection Techniques for Virtual Human Interaction

Virtual community simulation has been engaged on extensive research studies with rapid usage of virtual worlds. Realistic interaction with virtual humans is a complex research area in such kind of simulations since significant concern need to be devoted in order to simulate the social interactive phenomena. This research proposes a selection process for user interested virtual humans. The implemented selection techniques proposed to overcome limitations in existing virtual object selection techniques such as memory overhead and cognitive effort over the users. System prototype is integrated with multimodal interfaces to provide an immersive user experience in human computer interaction. Studying of the users has been conducted to comparatively evaluate proposed selection techniques with ray casting metaphor which is widely presented in games and simulations. Experiment has been conducted in order to evaluate both ideal and walking states of virtual humans. Results of the user study shows the suitability of proposed head oriented metaphors to interact in simulations which are related with virtual humans.

Simulating Gait and Structural Effects of Aging for Improved Diversity in Virtual Crowds

In real human crowds, diversification is caused by many factors such as gender, age, skin color, clothing, walking style, psychology etc., yet most crowd simulation systems rely only on appearance related variables for diversification. This research explores improving crowd diversity by simulating the biomechanical effects of aging in humans within crowd characters. Simulated characters were reprogrammed to be able to predict their behavior through an aging parameter predictor using support vector regression respective to their age. Data from gerontology studies recording spinal curvature and walking pattern deterioration with age were used for training. A spatial diversity algorithm was designed and implemented to distribute the resulting variants evenly among each other. Evaluations were conducted to gauge the impact on the diversification of the crowd simulation. The results prove that simulating posture and walking pattern deterioration in older characters does significantly improve biomechanical diversity within a virtual human crowd.

Machine Learning Based QSAR for Discovering Potential Drug Candidate from Endemic Plants of Sri Lanka - Case Study: HIV-1 RT

Diseases have become an inevitable aspect in the current world and the development of medicines have become a very important and crucial field. Computer-aided drug designing is an approach that is being used in the drug development process to minimize the high costs and time involved. An important method used along with this is QSAR where the mathematical relationships between chemical and molecular structures are identified. Plants have been proved to be a valuable and useful medicine for curing various kinds of diseases and Sri Lanka is a tropical country known as a country where Ayurvedic medicines and treatments are available. There are also a number of plants which are endemic to the country and could carry possible cures for diseases. The research presented in the paper aims to develop a model using a machine learning approach to identify potential drug candidates from a set of endemic plants of Sri Lanka.